Machine-driven classification of static objects, such as web pages, text documents, or multimedia content, is useful in contexts such as security and productivity.